The relationship between accelerometer and GPS derived data in elite men’s hockey competition

The advancements in micro-technology have enabled the quantification of external load in many team sports using GPS derived data. However, it has been proposed that an accelerometer derived variable, such as player load (PL), may provide a superior measurement system (Boyd, Ball, & Aughey, 2011, International Journal of Sports Physiology and Performance, 6, 311-321). The aim of this study was to examine the relationship between PL and distance covered in hockey competition. With institutional ethics approval, 25 elite male hockey players wore tracking devices in 8 international matches. Players were classified into 3 positional groups (defenders, midfielders and forwards). Absolute and relative values were determined for PL and distance. The Pearson product-moment correlation coefficient (r) was calculated to determine the relationship between PL and distance. The Fisher Z test was used to compare the relationships between groups. There were very large correlations between absolute distance and load for all positions (r = 0.902-0.956, P <0.01) and overall (r = 0.910, P <0.01). The absolute distance to load relationship was not different between positions. There were very large correlations between relative distance and load for all positions (r = 0.702-0.927, P <0.01) and overall (r = 0.872, P <0.01). The relative distance to load relationship was weaker in defenders compared with midfielders (Z = 3.48, P = 0.0005) and forwards (Z = 2.14, P = 0.0324). The strong relationships between these variables suggest that PL is simply a proxy for distance covered and it appears to be heavily influenced by the vertical accelerations that occur during foot strike at any running speed. Therefore, additional research is required to examine how PL can be modified to fully represent all hockey specific activities. The numerical spread of data may have influenced the positional relationships for the relative values.